From owner-freebsd-bugs Fri May 9 02:10:04 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id CAA22184 for bugs-outgoing; Fri, 9 May 1997 02:10:04 -0700 (PDT) Received: (from gnats@localhost) by hub.freebsd.org (8.8.5/8.8.5) id CAA22170; Fri, 9 May 1997 02:10:02 -0700 (PDT) Date: Fri, 9 May 1997 02:10:02 -0700 (PDT) Message-Id: <199705090910.CAA22170@hub.freebsd.org> To: freebsd-bugs Cc: From: j@uriah.heep.sax.de (J Wunsch) Subject: Re: gnu/3553: the -L option of tar does not work properly with long tape lengths Reply-To: j@uriah.heep.sax.de (J Wunsch) Sender: owner-bugs@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk The following reply was made to PR gnu/3553; it has been noted by GNATS. From: j@uriah.heep.sax.de (J Wunsch) To: aloke@vicor-nb.com Cc: freebsd-gnats-submit@FreeBSD.ORG Subject: Re: gnu/3553: the -L option of tar does not work properly with long tape lengths Date: Fri, 9 May 1997 10:32:00 +0200 As aloke@vicor-nb.com wrote: > GNU tar version 1.11.2. The -L option to tar is supposed to specify > the length of the tape in K, but when the number is large (I was > using 3145000 to specify a 3Gig tape) the user is prompted to mount > a second volume too soon. A look at the source shows the tape_length > variable to be a signed int, instead of unsigned. That won't help very much. Bumping the limit from 2 GB to 4 GB probably ain't worth the while. Either, it should be made a 64-bit entity at all (but that's probably quite a little more of work), or it's too small for many current tape drives already. Silly question: why do you want to specify -L at all? It is only required for broken hardware/drivers. The floppy disk driver used to be broken until FreeBSD 2.2 and required it, but to the best of my knowledge, the tape drivers were always able to correctly report EOF back to the caller. -- cheers, J"org joerg_wunsch@uriah.heep.sax.de -- http://www.sax.de/~joerg/ -- NIC: JW11-RIPE Never trust an operating system you don't have sources for. ;-)